Page 1

The Evolution of Citizenship Economic and Institutional Determinants ∗
Page 2

ABSTRACT We investigate the evolution of the legal institution of citizenship from a political economy perspective. We first present a median voter model of citizenship laws determination. Next we test the implications of the model on a new dataset on citizenship laws across countries of the world. We show that citizenship laws have responded endogenously to economic and institutional determinants. When facing increasing immigration, countries with a jus soli tradition tend to restrict their legislation, while jus sanguinis countries resist innovation. The welfare burden proves not to be an obstacle for jus soli legislation, while demographic stagnation encourages it. A high degree of democracy promotes the adoption of jus soli elements, while the instability of state borders determined by decolonization impedes it. Religion and ethnic diversity have no residual impact. .Key Words: citizenship laws, international migration, legal origins, democracy, borders. 1
Page 3

1. Introduction Each country of the world has established a complex system of rules that govern the attribution of citizenship. As a consequence of the increasing pressure of international migration, citizenship laws have moved to center stage on policy agendas, since citizenship laws not only affect the design of immigration policy, but also interact with the workings of labor markets, affect welfare programs, and influence demographic trends. Citizenship is the legal institution that designates full membership in a state and the associated rights and duties. It provides benefits such as the right to vote, better employment opportunities, and the ability to travel without restrictions, legal protection in case of criminal charges, and the possibility to obtain a visa for a relative. There are also costs to citizenship, such as the military draft, renunciation of the original citizenship, and the pecuniary and non-pecuniary costs that may be required for naturalization and for recognition at the age of majority. Examples are language and culture tests, waiting periods, and a commitment to avoid activities leading to disqualification. There are several ways to acquire citizenship: at birth, by naturalization, by marriage.

The regulation of citizenship at birth, which determines citizenship acquisition by second- generation immigrants, is rooted in the well-defined bodies of common and civil law. The former traditionally applies the jus soli principle, according to which citizenship is attributed by birthplace: this implies that the child of an immigrant is a citizen, as long as he is born in the country of immigration. The latter applies the jus sanguinis principle, which attributes citizenship by descent, so that a child inherits citizenship from his parents, independently of where he is born. Despite being rooted in these principles, during the 20th century - and especially after World War II - in many countries citizenship laws have gone through a process of continuous adaptation, in conjunction with the decolonization phase, the collapse of the socialist system, and the mounting pressure of international migration. In this paper we investigate the determinants and evolution of citizenship laws in the postwar period from a political economy perspective. To pursue this goal, we assemble a new dataset which codifies citizenship laws across the countries of the world, with a specific focus on the provisions that regulate the access to citizenship at birth. The dataset is then used to study the dynamic adaptation of these laws, by relating the observed patterns to a
Page 4

number of potential determinants, including economic factors as well as other political and cultural factors which have been found relevant in related research on institutions. Modern sociopolitical theories have advanced several hypotheses concerning the determinants of citizenship laws dynamics, on the basis of case studies and nonquantitative cross country comparisons. The legal tradition established in a given country is generally believed to exert a persistent impact on current legislation. The relevance of migration has also been investigated. In particular, pressure from a large stock of migrants is perceived as a factor that shapes a country's attitude toward citizenship policy. On the one hand, it could push toward a legislation that allows automatic citizenship granting for all newborn. On the other, migration could also drive toward restrictions of the same principle in countries where it was originally applied. According to some sociopolitical theories, the combination of these forces tends to induce convergence toward a mix of jus soli and sanguinis provisions for countries coming from different legal traditions (Weil 2001). For the case of Europe, Baubock et al. (2006) instead stress the presence of divergent trends, towards liberalization in some countries and toward restriction of access to citizenship in others. The influence of other economic forces is also recognized. Since citizenship rights determine the ability to enjoy welfare benefits, the shaping of nationality laws has been linked with the nature of the welfare state, with a large government representing a potential obstacle to the retention of jus soli (Joppke 1998). This argument, however, has to be weighted against the potential gain coming from the acquisition of relatively young new citizens for countries with expensive

pension systems and in the midst of a demographic crisis. Political factors have also been found relevant. The presence of a consolidated democracy is expected to lead to the adoption of jus soli, viewed as a more equal treatment of aliens. Stabilization of state borders should reduce the pressure to preserve a national identity through jus sanguinis. Finally, an additional factor that has been the subject of debate is the influence of national character and culture. The theory advanced by Brubaker (1992) focuses on France and Germany as two antagonistic kinds of nationhood, the former more assimilationists, and the latter more ethnocentric, which also differ in their definitions of citizenship. In this paper we formalize the above hypotheses within a simple median voter model which can guide our empirical investigation by generating testable implications and by offering an interpretation of the resulting evidence.
Page 5

The model is based on the assumption that the laws regulating citizenship acquisition can be viewed as the outcome of the decision problem faced by a native median voter, in a context where citizenship confers the right to vote over policy variables. In making this decision, the native median voter takes into account the associated benefits and costs, which depend on the share of migrants over population and the other factors suggested by the literature. The model indicates that the share of migrants has a potentially ambiguous impact on the decision to grant citizenship and voting rights to migrants, since it increases the loss, for the native median voter, of letting migrants vote, but it also increases the cost associated with their exclusion. Moreover, the natives' decision is positively influenced by a relatively high income level of the migrants, a small welfare state, a relatively old native population, a high level of democracy, a stable national border, and an inclusive national culture. We test the above empirical implications and find that in the postwar period citizenship laws have responded endogenously and systematically, through a slow but steady process of adaptation, to economic and institutional factors. Overall, our results suggest that migration pushes national legislations in the direction of jus sanguinis, not jus soli. In particular, when we take into account the legal tradition established in the matter of citizenship, we find that countries with a jus soli origin react to increasing migration by adding jus sanguinis elements. On the other hand, in jus sanguinis countries the impact of migration has been negligible. Therefore, the evidence does not support the hypothesis of convergence toward a mixed regime, since migration tends to induce restrictions, but not extensions. Other economic factors also matter. While the welfare burden proves not to be an obstacle for a jus soli legislation, demographic stagnation encourages the adoption of mixed and jus soli regimes.

Turning to institutional factors, we find that a high degree of democracy is significantly associated with a jus soli legislation while border instability, in particular following the decolonization phase, decreases its likelihood. Cultural characteristics captured by religion and ethic fractionalization are not found to play a significant role. The rest of the paper is organized as follows: Section 2 introduces the related literature. Section 3 reviews the historical and legal background for the issues we address. Section 4 presents our model of citizenship laws determination. Section 5 describes our dataset on citizenship laws around the world 4

Page 6

Section 6 investigates empirically the determinants of current citizenship laws and presents our main results, together with a set of robustness checks. Section 7 develops an alternative empirical strategy that highlights the determinants of change in citizenship laws. Section 8 concludes and indicates directions for future research. The Data Appendix collects information about the data employed. 2. Related Literature Our work is related with several branches of the economic literature. First of all, this paper adds to research on international migration and migration policy. Timmer and Williamson (1998), Hatton and Williamson (2006) and Bertocchi and Strozzi (2008) empirically analyze immigration policies enacted at the end of the 19th century during the mass migration era, while O'Rourke and Sinnott (2006) and Mayda (2006) estimate voters' attitudes toward immigration in the postwar period. The political economy of migration has been modeled, among others, by Benhabib (1996), Gradstein and Schiff (2006), and DeVoretz (2006). More specifically, the role of citizenship policy has been discussed by DeVoretz and Pivnenko (2006), who investigate the economic costs and benefits derived from citizenship, by Pritchett (2006), who evaluates citizenship policy within a broader discussion on labor mobility and

We add to this stream by focusing on the determinants of the dynamic adaptation of nationality rules. the issue of franchise extension has recently received considerable attention within the literature. On the other hand. our work is also related to Razin. and Robinson (2001) contribute to the understanding of how institutions evolve by using historical variables as instruments for contemporary measures of the quality of institutions. However. (1998). More broadly. . This paper also relates to the comparative legal approach initiated by La Porta et al. The basic premise of this research line is the recognition that laws in different countries are adopted or transplanted from a few legal traditions and that the resulting legislative bodies reflect both the influence of the legal origin and the subsequent revision specific to individual countries..immigration policies. but with a focus on the conflict between rich and poor. while Bertocchi (2007) concentrates on the conflict between men and women. our work contributes to the research program which has focused on the historical determinants of institutions. and by Chiswick and Miller (2008). Sadka. Balas et al. and Swagel (2002) and Dolmas and Huffman (2004). they do not examine the determinants of these alternative regimes. who compare the impact of migration on the welfare state with or without voting rights for the migrants. Acemoglu. Johnson. (2008) specifically analyze the evolution of an index of formalism of legal procedure. Since our theory emphasizes that citizenship rights imply the right to vote. Engerman and Sokoloff (2002) highlight the relevance of wealth inequality and political factors in accounting for how fundamental economic 5 Page 7 institutions developed over time. Bertocchi and Spagat (2001). who evaluate the determinants of the decision to naturalize. with contributions by Acemoglu and Robinson (2000).

. Continental modern citizenship law was subsequently built on these premises. At the same time. In 18th century Europe jus soli was the dominant criterion. and thus state borders. recent work by Alesina and Spolaore (1997) and Bolton and Roland (1997) on the optimal determination of the size of nations. Japan also adopted jus sanguinis in this phase. By imitation. During the 19th century the jus sanguinis principle was adopted throughout Europe and then transplanted to its colonies. that are typically adjusted to the business cycle and to the current government orientation. By the end of the 19th century. both because country size in this literature is the same as population size and is potentially influenced by migration and by the legal status of immigrants. following feudal traditions which linked human beings to the lord who held the land where they were born. the British preserved their jus soli tradition and spread it through their own colonies. contrary to other migration policy measures such as quotas and visa requirements. Citizenship Laws in Historical Perspective Citizenship policy can be viewed as part of broader migration policy. starting with the United States where it was later encoded in 6 Page 8 the Constitution. the process of nation-state formation and the associated codification effort were completed in Continental Europe. 3. However. and also because borders play an important role on the determination of citizenship rules.Finally. is also relevant to our approach. On the other hand. The French Revolution broke with this heritage and with the 1804 civil code reintroduced the ancient Roman custom of jus sanguinis. citizenship laws reforms tend to be the outcome of long-term processes of adaptation often involving constitutional amendments.

drawing mostly from Joppke (1998). while civil law France. with the specific purpose to protect the birthrights of black slaves. 2 Australia Current citizenship law in Australia differs considerably from that of the United . with jus soli being the norm in common law countries. For instance. had by then already moved toward a mixed regime. However. 1 In particular. and 19th century colonization had extended the process of transplantation of legal tradition to the rest of the world. Therefore. Aleinikoff and Klusmeyer (2000. but never led to actual change. and with a general positive attitude toward economic liberalism. civil law Latin America had embraced jus soli early on. 2001) and Brubaker (1992). the US approach is still remarkably consistent with its original attitude in all its aspects. Consistently with its history as a country of immigrants. despite important exceptions. most countries of the world had established specific provisions regarding citizenship acquisition within a relatively well-developed legal system. A relatively young and thin welfare state contributes to the fiscal sustainability of jus soli in this country. the next century witnessed a continuous process of transformation of citizenship laws across the world. jus soli came under attack in the 1980s regarding its applicability to the children of illegal immigrants. ranging from immigration policy to naturalization requirements. and jus sanguinis regulating citizenship law in most civil law countries.the revolutionary phase was over in those countries that had been the subject of the earlier colonization era. Below are some specific cases. with its colonies. Debate about possible restrictions did arise recently. by that stage. The United States Jus soli was encoded in the US Constitution through the 1868 Fourteenth Amendment.

2 The relative thickness of the concept itself of citizenship. Pritchett (2006) discusses the possible advantages of guest-worker programs which do not contemplate citizenship. potentially relevant consideration: in the US. Therefore. Latin America In the face of a civil law tradition which had been transplanted by the European powers. of Argentina in 1853. 7 Page 9 States. Mexico represents a special case where jus soli was also . Jus soli survived until 1986. citizenship is relatively thin. from the perspective of poor countries.1 In his analysis of Mexican immigration. is a related. most of Latin America was already a jus soli country before the 19th century immigration waves began. while afterwards a person born in Australia must have at least one parent who is either an Australian citizen or a permanent resident in order to acquire citizenship. Jus soli had also been introduced in Australia by the colonists. Jus soli was encoded in the Constitution of Brazil in 1824. this area has followed a rather peculiar pattern. Jus soli is still the prevalent rule in the area. in the sense that it confers few additional benefits if compared with residency. Huntington (2004) has criticized current nationality regulations on the grounds that they represent a “devaluation of citizenship”. of Venezuela in 1830. for instance. despite the common origin as countries of immigration. even if it is no longer attracting immigrants. the country went through numerous legislative and administrative reforms. if compared to residency. most of the incipient states chose jus soli as a way to break with the colonial political order and to prevent the metropoles from making legitimate claims on citizens born in the new countries. In the postwar period. At independence.

France The emergence of the nation-state in Continental Europe was the main factor that shaped citizenship law in this area. In order to secure immigrants' children born in France to the draft. the concept of nationality in the UK was. in 1889 double jus soli became automatic. raised concern regarding assimilation. up to World War II. K. Citizenship issues and the rights of immigrants became the object . where jus sanguinis was first introduced with the 1804 Civil Code and maintained 8 Page 10 for the entire course of the 19th century. Following a postwar wave of colonial immigration. Because of its colonial history. especially from North Africa. Redefinitions of national citizenship have been effectively employed. qualifies for British citizenship only if at least one parents is a British citizen or resident.adopted in the 1814 insurgent Constitution. even though military consideration introduced early on elements of jus soli. The 1948 Nationality Act created the status of Citizen of the United Kingdom and Colonies for people with a close connection to the UK and its colonies. large-scale immigration. this open-door policy was progressively restricted. After World War II. since the 1980s. even though special status is still attributed to citizens of the British Commonwealth. as a form of selective immigration policy. The United Kingdom British nationality law has been deeply affected by the imperial experience. The revolutionary experience was particularly important for France. making the experience of this country a unique one. The 1984 British Nationality Act restricts jus soli by establishing that a child born in the U. since all subjects of the British Empire had equal access to British citizenship simply by establishing residence in the UK. only to come back to stay with a Constitutional Amendment in 1937. particularly extensive. but was then abandoned in 1836.

A major overhaul of the legislation. A first step in this direction was the new Foreigner Law in 1990. Brubaker (2002) has influentially argued that the different path followed by these countries has been shaped by their cultural difference. which had established strong sanguinis ties with German overseas emigrants. but to no avail.of heated debate in French politics. The case of France is frequently compared with Germany. the original Wilhelminian citizenship law of 1913. With the Left regaining political power in 1997. With the foundation of the GDR and the consolidation of the Eastern Block. however. and at the same time with millions of ethnic Germans living behind the Iron Curtain. had started to put under strong pressure. mostly from Turkey but also from Southern Europe. which turned naturalization from the discretionary exception into the rule. that required a formal citizenship request from second-generation immigrants. and Germany to its ethnic identity. the massive guest-worker immigration of the postwar period. Achieving border stability was a decisive factor in pushing Germany toward the longdelayed adoption of jus soli elements. with France sticking to its tradition of assimilationist nation. which paved the way for the achievement of stable national borders. Germany The single most relevant event in the history of German citizenship law is certainly the fall of the Berlin wall. In 1993 Chirac introduced a restrictive revision to the legislation. Germany found itself in the paradoxical situation of having to live with a large population of disenfranchised foreigners born on its soil at home. Prior to that. nor declined it. following an intense political struggle. was finally approved in . these restrictions were considerably revised. with the automatic assignment of citizenship at age 18 to those immigrants' children born in France who had neither requested.

which recently adapted their legislation to the globalization of international migration and its increasing impact on Europe. many former British and Portuguese colonies rejected the jus soli tradition and switched to an often strongly ethnically-tinged version of jus sanguinis. but had recently to adapt to the quickly changing conditions. As documented by Weil (2001). in the vast majority of European countries. and not only through the indirect impact on the metropolitan countries we previously examined. especially for high-immigration Sweden. the strong ethnic character of German national identity. as mentioned in the introduction. where jus sanguinis was functional to the large past emigration flows. . other factors that may have delayed the introduction of jus soli are.1999. The vast majority of the African colonies that were subject to civil law countries practicing jus sanguinis stuck to this principle after independence. In the evaluation of the German experience. with the exceptions of Austria. restricted forms of double jus soli are de facto applied. For instance. by now. On the other hand. The latter aspect may have played a role in shaping the evolution of citizenship policies in several other European countries and especially the Scandinavian ones. Decolonization Postwar decolonization had a major impact on citizenship rules applied around the world. in the entire EU. In particular. Greece and Luxembourg. access to citizenship by second and third generation is facilitated. Jus soli is now the norm in Germany (under the mild requirement that one parent has lived in the country for eight years). and the thick nature of the 9 Page 11 German welfare state.

sizeable minorities . The Soviet Union had occupied Estonia. the associated exclusive notion of ethnic and tribal identity caused enormous problems in countries where colonial rule had left shaky democratic institutions. 10 Page 12 The disintegration of the USSR Another major wave of citizenship law codification followed the disintegration of the USSR.Sierra Leone's 1961 Constitution established that citizenship is transmitted only by descent and only to children whose father and a grandfather were Sierra Leoneans of AfricanNegro descent. ethnic conflict lies at the roots of a chronic manipulation of citizenship rules in favor of one ethnic group over others. there had been considerable migration within. however. During the following decades millions of Russians were encouraged to settle in Latvia and Estonia (less so in Lithuania) in order to Russify them. stateless. in an effort to exclude Rwandan immigrants. and was compounded with deep ethnic division. as for all empires. In 1981 Mobutu signed a new law on nationality requiring an ancestral connection to the population residing in the territory as far back as 1885. recognized citizenship only for persons whose parents were members of one of the tribes established within the territory by 1908. Latvia and Lithuania in 1940. In situations where instability was pushed to an extreme degree by the young age and the arbitrary borders of these countries. To these days. jus sanguinis tended to prevail as a way to control more easily the formation of national entities. large Russian-speaking. Marginalization and de facto statelessness of significant strata of the population is the unavoidable outcome of these policies. The 1964 Congolese Constitution. At the same time. The area had been sealed toward international migration but. To these days.

We can view the laws regulating citizenship acquisition as the outcome of the decision problem faced by a median voter. this perception as a country of emigrants pushes toward the persistence of jus sanguinis as the main principle. while Lithuania. showed a more open approach. Estonian and Latvian laws were sharply criticized by international organizations on the grounds of human rights.are still present. in a context where citizenship confers the right to vote . the new citizenship laws of these three states reflected this heritage with an emphasis on jus sanguinis as the basis for acquiring citizenship. The hostile attitude toward ethnic Russians was especially strong in Latvia. the salient fact in shaping current citizenship policy is the perception that many of its citizens are outside its borders. which was less affected by Soviet immigration policy. even though small concessions to jus soli have been made. 4. these recommendations were indeed fulfilled in the more recent legislation of the Baltics. After independence. The Model This section develops a simple theoretical model which formalizes the hypotheses that a widely interdisciplinary literature has advanced with respect to the determinants of citizenship laws. while most other countries of the area still persist with discriminatory policies. and a commitment to democratic values with respect to the rights of minorities. for the case of the Russian Federation. The issue for these states was how to balance a need to reconstitute their national identity around an ethnic model. The aim of the model is also to generate testable implications that can guide our interpretation of the empirical evidence. Again. In the anticipation of EU integration. spread around the former regions of the USSR. By contrast.

y N >y>y M . even if the model concentrates on voting on a specific policy. namely. and 11 Page 13 thus the right to vote. the same approach could be extended to consider alternative agenda. We consider an economy where a population of mass P consists of natives with mass N and migrants with mass M. beside the right to vote. and y = N P y . respectively. Finally. a redistributive tax scheme which finances a public good. Migrants are poorer than natives since they are relatively unskilled. namely. and also implies some duties. in our oneperiod framework the distinction between different ways to acquire citizenship becomes irrelevant. A few warnings are in order. when considering the decision to grant citizenship. First. to migrants. where M + N = P and M < N. A native voter. economic and social goals.over policy variables. our focus on the right to vote as the main benefit of citizenship is easily justified. therefore the model's predictions can be applied to the laws concerning both citizenship at birth and naturalization. Second. since political rights can be viewed as an instrument through which migrants could achieve broader political. before we present the model's details. is driven by the benefits and costs associated with this decision. while in practice citizenship acquisition implies a larger set of rights. where y N and y M denote average income for natives and migrants.

We also assume that income distribution is skewed to the right for each group. namely.N + M P y M is the economy-wide average income. as in Meltzer and Richard (1981). according to a proportional income tax rate τ. Both natives and migrants derive utility from consumption of a private good. g. according to u i =c i + λg (1) where λ is a positive preference parameter. Both groups pay taxes. c i . Each enfranchised individual casts a vote on the tax rate. and thus for the economy as a whole. Assume initially 12 Page 14 . and a public good. such that 0 < τ < 1. The tax rate is set through a political choice under majority voting. Tax revenues are used by the government to finance the public good according to the following balanced budget constraint: g = τy − τ 2 2 y (2) where the second term captures tax collection costs. median income is lower than average income both for natives and migrants.

according to k=K+h M P (3) where K reflects the degree of cultural inclusiveness and h > 0. Equivalently. k could enter directly the utility function. as captured. by a jus soli tradition. The cost is also affected by factors that determine the degree of inclusiveness of a country's culture. reflecting the possibility that their disenfranchisement can lead to social unrest and even violence.that only natives are citizens and are therefore allowed to vote. for instance. Assume also that society bears a cost k for the exclusion of migrants from citizenship. The expression for the indirect utility function of a native voter with income y N i is given by v N i = (1 − τ)y N i − k + λ(τ − τ 2 2 . This cost increases with the share of migrants over population. The cost enters the individual budget constraint as follows: c i ≤ (1 − τ)y i −k (4) where y i denotes individual income.

)y (5) which is single-peaked with respect to the tax rate. To be noticed is that under our assumptions about income distribution it is not necessarily the case that y N∗ y < 1.which implies that the tax rate is going to be positive only if y N∗ y < λ. which is given by . since migrants also pay taxes. according to which the equilibrium tax rate is the preferred tax rate of the native median voter with income y N ∗ . The native median voter could avoid the cost k by granting citizenship to the migrants and thus accepting the tax rate that would prevail under universal enfranchisement. which is measured by the ratio of native median income over the economywide average income. according to τ N ∗ =1− 1 λ y N ∗ y (6) The level of the tax rate increases with the intensity of the preference for public goods and with inequality. We can therefore apply the median voter theorem.

In particular. the latter is certainly positive. Since y N ∗ >y ∗ . but can enjoy a smaller tax. it follows that τ N ∗ <τ ∗ . The native median voter faces a simple set of costs and benefits when considering the decision to grant citizenship. he has to pay more taxes but k is avoided. The difference between τ N ∗ and τ ∗ increases with the income gap between natives and migrants and with the share of migrants over population. namely. the tax rate chosen by the native median voter is lower than the tax that applies under universal suffrage. If migrants cannot vote. given our assumptions on income distribution. he has to face the cost k.τ ∗ =1− 1 λ y ∗ y (7) 13 Page 15 where y ∗ is the economy-wide median income. It follows that the native median voter decides to grant citizenship to migrants if and only if . If migrants vote.

we can think of its dynamic implications in terms of a sequence . Both the disenfranchisement cost and the fiscal gain associated with the no-franchise status quo increase in the share of migrants over the population. Even if the model is static.(1 − τ N ∗ )y N ∗ − k + λ(τ N ∗ − τ N ∗ 2 2 )y ≤ (1 − τ ∗ )y N ∗ + λ(τ ∗ − τ ∗2 2 )y (8) where following (5) we find on the left hand side his indirect utility function when migrants cannot vote and on the right hand side his indirect utility function when migrants can vote. while the cost increases with the degree of inclusiveness K of the country's culture. Besides. the fiscal gain also increases with the income gap between natives and migrants.

While the present formulation of the model is designed to establish conditions for extension of citizenship rights to migrants. Together with the ethnic natives. it follows that jus sanguinis countries will be more reluctant to change. Finally. If. it can also encompass restriction. following a sequence of stationary decisions. migrants who are already in the country and have thus become citizens are simply to be considered as natives themselves. When the status quo is a jus soli regulation.of repeated decisions. since it is the share of migrants over population that matters. they will decide whether or not restricting the current regulation taking into account the incoming waves of immigrants and following the simple logic previously illustrated. To be noticed is that a given stock of migrants has a stronger impact on countries with a relatively small native population. the economy is shocked by an increase of the migrants share. due to a large inflow. the effect of an increase in the share of migrants is potentially ambiguous. If we interpret a jus sanguinis tradition as a low degree of inclusiveness and thus a low cost of exclusion. The predictions so far obtained from the model indicate that a decision to extend citizenship and the associated voting rights is facilitated by a smaller income gap between natives and migrants and by a larger degree of inclusiveness of a country's culture. generating a trade off. the median voter will respond taking into considerations all the channels involved and this may result in an adaptation of the regulation. . The net effect will depend on which factor is stronger. or equivalently a particularly generous naturalization policy. since a higher share increases 14 Page 16 both the cost and the fiscal gain of disenfranchisement.

Empirically. While our one-period model cannot explicitly reflect these aspects. which can be modelled with an income franchise requirement. border instability could be captured in a version of the model where the size of the native population. If only rich natives are allowed to vote. is subject to uncertainty. Fourth. and thus the population share of migrants. the level of democracy can influence the outcome since it implies a constraint on the political rights of the natives themselves. This should facilitate the decision to grant them citizenship and implies that countries with a relatively old native population should be particularly sensitive to these considerations and thus display a more open attitude. as captured by λ. by increasing the tax differential.The basic model can be extended to consider several other potentially relevant factors. Third. thus amplifying the tax cost which follows the decision to allow migrants to vote. the impact of the size of government on citizenship laws can be captured by assuming that different countries exhibit different preference parameters toward government. Second. 15 Page 17 . The testable implication is that the decision to grant citizenship is positively influenced by the domestic level of democracy. which is higher for a migrant. we can interpret our tax as a life-long contribution. at any given level of income. we should therefore expect a negative impact of the size of government on the degree of inclusiveness of citizenship laws. Thus a relatively large government size. could make an open citizenship policy more costly. First. demographic aspects could be considered by assuming that the migrants' younger average age implies a larger ability to contribute to the welfare state. the associated tax rate will be higher than otherwise.

of letting migrants vote. even though there were nearly no reforms in citizenship laws during the first half of the century. and the Department of State. the United Nations High Commission for Refugees (2003). The Data: Citizenship Laws of the World We compile a dataset of citizenship laws across the countries of the world for the postwar period. which provides synopses of the citizenship laws currently practiced in 190 countries. and a large native population. Citizenship at Birth We attribute to each country an appropriate code for citizenship laws in 2001. to introduce a random component in the model should affect the voting decision by reducing the tax rate. but it also increases the cost associated with their exclusion. the main problem is to establish who can be considered as a native. a high level of democracy.Indeed. a relatively old native population. we expect the natives' decision to be positively influenced by an inclusive national culture. a relatively high income level of the migrants.1. 1975 and at the beginning of the postwar period. the Library of Congress. The principal focus of our codification is citizenship acquisition at birth. a small welfare state. for the native median voter. Under standard risk aversion behavior. We supplement this information with additional one from the CIA World Factbook (2002). the share of migrants has a potentially ambiguous impact on the natives' decision to extend citizenship and voting right to migrants. 5. but we also collect information about naturalization requirements. The principal source for the information we codify is a directory published by the Investigations Service of the United States Office of Personnel Management in 2001. a stable national border. The sources for this directory were Embassies. and the survey in Weil (2001). when the national border is unstable. We take 1948 as the starting point. Moreover. To summarize. since it increases the tax disadvantage. 5. so that .

the subsequent half century did not see further evolution. which gained independence from the British and French administration in the 1943-1948 period. we divide countries into three groups: countries subject to jus sanguinis without any jus soli element (Group 1). 1975 and 2001. common law tradition. and countries subject to full jus soli (Group 3). despite the occurrence of major historical events such as World War I. For 1948. A mixed regime includes elements of both jus soli and jus sanguinis. 4 Our dataset includes those 162 countries for which we were able to collect information on both original . rather than emigrants. countries that apply a mixed regime (Group 2). 3 As in 3 By treating 1948 as the initial year. Indeed.most of the legislation in place in 1948 had actually been developed much earlier. at least relative to the subsequent developments that are the focus of the present investigation. This approach is justified by our primary interest in the potential impact of citizenship laws on immigrants. 16 Page 18 the analysis of legal origins in La Porta et al. (1998). we include the postwar decolonization phase with the exemption of the Middle East. we treat the specific legal provisions regulating access to citizenship in 1948 as predetermined. In our classification we focus on the presence of jus soli elements in a country's legislation. while the 19th century witnessed a first wave of adaptation of citizenship legislation from the civil vs. we divide the postwar period into two subperiods of equal length. By coding citizenship laws in the intermediate year 1975.

and current citizenship laws, and for which migration data were available for the postwar period. 5 INSERT TABLE 1 The differential patterns of evolution that citizenship laws generate in 1948, 1975 and 2001 are summarized by the transition matrices in Table 1, which reveals considerable variations both across countries and over time. The table shows that in 1948 jus soli was the rule in about 47% (namely, 76 out of 162) of the countries, while jus sanguinis dominated in 41% (namely, 67 out of 162), and the mixed regime was adopted in the remaining 12% (19 countries). Among the countries that were under jus soli in 1948, we find the United States, Canada, all the Oceanian countries, most of Latin America, within Africa and Asia the British and Portuguese colonies, within Europe the UK, Ireland and Portugal. On the other hand, in 1948 jus sanguinis predominated in most of Europe, including its Eastern part. As explained in Section 3, France was unique in its early choice of a mixed regime. Since we 4 For details on our classification criteria see the Data Appendix, part A. 5 For details on migration data see the Data Appendix, part C. 17
Page 19

treat colonial territories as subject to the metropolitan countries' regime until independence, the group applying the mixed regime in 1948 includes France and its colonies. By 1975, 31% (namely, 50 out of 162) of the countries had jus soli, 62% (101) jus sanguinis, and 7% (11) a mixed regime. The main event justifying this evolution is decolonization, with many former colonies switching to jus sanguinis, from jus soli when the UK and Portugal

were the metropolitan country, and from the mixed regime in the case of France (see Section 3). As of 2001, 24% (namely, 39 out of 162) of the countries apply jus soli, 54% (88) jus sanguinis, and 22% (35) a mixed regime. It has mostly been the adaptation of the legislation of many European countries, relaxing pure jus sanguinis in favor of a mixed regime, that explains the pattern observed for the second subperiod. Among the countries that still adhere to the jus soli principle in 2001 are the United States, Canada, New Zealand, and Ireland (which - however - recently introduced restrictions to jus soli with a June 2004 referendum). The United Kingdom and Australia, on the contrary, no longer adhere to it and now adopt a mixed regime. Overall, jus sanguinis is currently the most common regime, with 69% of the countries in Africa, 83% in Asia, and 41% (down from 88%) in Europe. The growing group where a mix of provisions is applied is particularly well-represented in Europe, with 56% of the European countries including the formerly jus soli United Kingdom. On the other hand, jus soli predominates in the Americas, with 89% of the countries in Latin America, and the entire North America (namely, the U. S. and Canada). Table 1 reveals three different patterns of transitional dynamics: stability, switch, and convergence. Stable countries lie along the diagonal. Looking at the 1948 to 2001 transition, we see that a large fraction (28%, namely, 46 out of 162) have started and ended as jus sanguinis. In other words, it is 69% (namely, 46 out of 67) of the originally jus sanguinis countries that have remained so. By contrast, 22% (36 out of 162) are steadily jus soli countries: this means that only 47% (36 out of 76) of the originally jus soli countries have not changed their policies. Off diagonal, there is a sizeable proportion of countries (19%, or 31 out of 162) that have switched from jus soli to sanguinis, by completely eliminating

birthplace as a criterion: most of them - as mentioned - are former African colonies of the UK and Portugal, which made this radical choice at independence. Looking at the two 18
Page 20

subperiods, we see most of these switches occur between 1948 and 1975. Finally, there is evidence of a process of convergence to a soli/sanguinis mix, which affects 18% of the countries (29 out of 162, of which 20 converge from jus sanguinis by adding jus soli elements, while 9 converge from jus soli by restricting it) and intensifies between 1975 and 2001. INSERT TABLE 2 In Table 2 we present further information on citizenship laws evolution by reporting changes in citizenship laws, organized by original laws. Over the 1948-2001 period, 74 countries (46%) have gone through a change in the laws. Of these, 51 have changed toward jus sanguinis and 23 toward jus soli, while 45 changes have occurred in the first subperiod and 33 in the second. 6 In particular, in the first subperiod, the majority of the countries that went through a change (29, or 64%) were originally jus soli. As mentioned, this pattern is determined largely by the behavior of former colonies. In the second subperiod, the majority of the countries that went through a change (20, or 61%) were originally jus sanguinis, most of which adopting a more open legislation. The above discussion suggests a relevant role of border stability. To investigate this issue, we introduce a set of dummies capturing a country's history of border changes. In particular, we distinguish across three different causes of border instability: decolonization, Berlin wall, other border changes. 7 If we compare the transitional dynamics of the full sample with those

of the subsample of countries that did not go through a border change, we count for the latter a much smaller proportion of switches from jus soli to jus sanguinis. This pattern confirms the relevance of border changes, especially those due to decolonization. INSERT TABLE 3 Summary statistics for our citizenship laws dataset are reported in Table 3. The correlation between 1948 and 2001 citizenship laws is 0.42, which points to some persistence, as confirmed by the even higher correlation between 1948 and 1975 (0.60) and 1975 and 2001 laws (0.81). 6 A few countries went through more than one change. 7 The Data Appendix, part B describes how the three border change dummies are constructed. 19
Page 21

5.2. Citizenship by Naturalization and the Citizenship Policy Index Naturalization policies are also relevant to the issues at hand. Indeed, to facilitate naturalization for immigrant parents may represent a substitute mechanism to attribute citizenship to children born in jus sanguinis countries. Besides, the general attitude revealed by a country's regulation of citizenship at birth may be reflected in its naturalization laws, with jus soli countries traditionally making naturalization much easier, at least for resident aliens. Within jus sanguinis countries, naturalization requirements again tend to be correlated with the revisions introduced for citizenship at birth. Basic rules for naturalization may include a period of residence, renunciation of other citizenship, familiarity with the language and customs of the country, and the availability of adequate means of support. We code naturalization only for 2001, on the basis of the available information on 142 of our 162 countries. We classify countries on the basis of the number of years of residence

We construct an index of citizenship policy defined on the 0-1 interval. while Baubock et al. which can be considered a relatively open attitude.54. even though data on naturalization are only available for the year 2001 and for a subset of countries. we treat citizenship laws in 2001 as an ordinal variable. but we do not use this variable due to limited information. We then combine the information we collected on citizenship at birth and naturalization within a single measure.required for naturalization. (2006) collect statistics on acquisition of nationality. the British Council has compiled an index of civic citizenship and inclusion (see British Council Brussels 2005). which can be explained by the fact that the correlation between citizenship laws and naturalization is only 0. 20 Page 22 . 5 years. 6 to 14 years. by constructing four classes (more than 14 years. 10 8 We do not consider naturalization by marriage. 44%) require five years of residence.37. 9 Dual citizenship provisions constitute another potentially relevant aspect of citizenship policy. 8 In our dataset 62 countries (namely. Alternative ways to define naturalization classes yielded similar conclusions. 10 For European countries only. 9 Table 3 reports summary statistics for naturalization and the citizenship policy index. 4 years or fewer). To construct the index. while 46% require more time and only 10% are more open. by associating jus soli elements with lower number of years of residence required for naturalization. The corrected Cronbach's alpha of the indicator is 0. since it is heavily dependent on family law.

which is categorical and can take three values: 1 if the country has a jus sanguinis regime. we also consider religion and ethnic diversity. In fact. we select the following variables.6. Among regressors. while we cannot gauge the impact of the income gap between natives and migrants due to lack of data. for example. The Determinants of Citizenship Laws 6.1. or else may not have any effect at all (and vice versa). this choice does not impose any ex ante ordering among the three regimes. and jus sanguinis with minimal inclusiveness. We also control for countries with a particularly small population size. and the age structure of the population. an ordering by increasing inclusiveness toward immigrants. in order to investigate the determinants of citizenship laws evolution in the postwar period. We interpret the presence of a jus soli tradition as an indicator of the degree of inclusiveness of national culture. in principle those regressors that affect the probability of being a jus soli country may not always have the opposite effect on the probability of being a jus sanguinis country. . we employ measures of migration. To control for other cultural characteristics. We use a multinomial logit specification to achieve maximum generality. Further discussion on this point is postponed to the sub-section on robustness. Moreover. Empirical Specification In bringing our theoretical model to the data. 2 if the country has a mixed regime. The empirical investigation is performed using a panel sample which includes information on two cross sections of 162 countries: the first cross section refers to the 1950-1975 subperiod. with jus soli being associated with maximal. and 3 if the country has a jus soli regime. The dependent variable is citizenship laws. the size of government. Border stability is measured through our border changes dummies.

162 and t = 1. M it S . L it represents citizenship laws in country i at the end of period t...2 (where t = 1 refers to the 1950-1975 period and t = 2 refers 21 Page 23 to the 1976-2000 period)...the second cross section to the 1976-2000 subperiod. (9) with i = 1. the multinomial logit model we run has the following form: L it = a + bM it + cS it + dM it S it + eT t +Z 0 it f+ it . 11 In specification (9). a is a constant term. In the full specification we present. S it is a dummy for the presence of jus sanguinis in country i at the beginning of each subperiod. M it is migration stock as a percentage of the population in country i at the beginning of period t.

where we take jus sanguinis as the reference category. Throughout the following. T t is a period dummy. and a measure of democracy. regarding the potential role of the above mentioned factors. The Data Appendix collects information on the definitions and sources of all variables we employed. The first group includes dummies capturing the country's geopolitical position. namely. starting with the variables we consider focal to our analysis. Table 4 presents their summary statistics. Within this group we consider the border changes dummies and dummies for Latin is the interaction between the previous two variables. INSERT TABLE 4 We can now suggest a number of specific hypotheses. consistent with the model previously outlined. 12 The set of explanatory variables Z it can be divided into two groups. we will organize our comments in terms of the effects of each of our regressors on the probability of adopting either a mixed or a jus soli regime instead of jus sanguinis. This follows the specification of our multinomial logit estimates. the share of young in the population. Southern European and small countries. Z it is a vector of additional explanatory variables. and it is an error term. proxies for cultural characteristics such as religious affiliation and ethnolinguistic fractionalization. Migration (M . The second group of explanatory variables includes the size of government as a share of GDP. immigration and legal tradition.

13 while for the second they refer to the stock in 1980. S i1 = 1 if country i has jus sanguinis in 1948 and S i2 = 1 if country i has jus sanguinis in in 1975. 12 In ) is measured by the stock of migrants in percent of the population at the beginning of each subperiod. respectively. By entering the migrant stock near the beginning of each period. L i1 and L i2 are citizenship laws in country i in 1975 and 2001. M i1 is migration stock in country i in 1960 and M i2 is migration stock in country i in 1980. since it is unlikely that stocks evaluated at the beginning of the period can be affected by subsequent changes in citizenship laws. T 1 = 0 and T 2 = 1. 14 A positive coefficient . 22 Page 24 1960. we avoid any potential endogeneity problem of migration with respect to citizenship laws. For the first subperiod. the available data refer to the stock in 11 We also run multinomial logit models for two types of more parsimonious specifications. for comparison purposes.

in the presence of a large stock of migrants. most of Latin America 13 . we should expect negative signs for the coefficients for our border change dummies. thus suggesting a particularly significant role of this initial legislation. In particular. The interaction between the jus sanguinis dummy and migration should reveal additional information: if positive. As explained in Section 3. thus confirming persistence of the original laws. A crucial control in our regressions. if border stability really counts as a prerequisite for the introduction of automatic birthrights for the immigrants. Turning to our geopolitical dummies. To assess the total impact of migration for jus sanguinis countries we also need to take into account the interaction's coefficients. We measure it with a dummy for countries that apply jus sanguinis at the beginning of each subperiod (S it ). as suggested by some of the political theories introduced in Section 1. We introduce the Latin America dummy to capture the peculiarity of this continent's experience. a positive coefficient for the mixed regime would suggest convergence toward the intermediate group. and similarly for the jus soli regime. is the legal tradition in the matter of citizenship. the coefficients of the interaction could also turn out to be negative since. the natives' reaction could be a conservative one.for the mixed regime would indicated that high migration pushes toward it rather than toward jus sanguinis. On the other hand. its coefficients would indicate that those jus sanguinis countries facing high migration tend to add jus soli elements. A negative value of the dummy's coefficient for a mixed and jus soli regime would imply that jus sanguinis countries are less likely to end up in the mixed and jus soli groups. We select this dummy because jus sanguinis is the most persistent of the three regimes. as previously discussed.

where alternative measures of migration are introduced. the endogeneity of our migration measure is ruled out by the fact that such changes had been extremely rare during the first half of the 20th century. 23 Page 25 adopted jus soli long before our sample period. since migration data reveal that countries with a small population tend to have large and erratic figures. see the Data Appendix. so its current position is not determined by postwar developments and in particular by its postwar migration experience. On the other hand. The size of government is meant to proxy for the nature of the welfare state: if a thicker. 14 Even taking into account the anticipation of future changes of citizenship laws in making migration decisions. we would find that countries with a higher share of young in . we should expect negative signs for this dummy's coefficients.Earlier data are not available. with a disproportionately small impact on their legislation. we should expect a positive coefficient for the mixed regime since these countries have been experiencing quickly increasing migration during the second subperiod. more expensive and more redistributive structure would represent an obstacle to automatic citizenship granting to the children of relatively poor immigrants.1. If indeed the behavior of Latin America differs significantly from the rest of the sample in being associated with a higher probability of adopting jus soli.3. The potential endogeneity of migration is further addressed in sub-sub-section 6. part C. Finally. it should exhibit a positive coefficient for this kind of legislation. we would found negative coefficients. with most of the revision to the legislation toward mixed regimes occurring in the past 15 years or so. if young immigrants could offer a solution to domestic demographic imbalances. For Southern Europe.

15). political rights.52). The share of young in population is positively associated with migration stocks (0. the small country dummy. the dependent variable. The dummy reflecting jus sanguinis as the initial law is negatively correlated with decolonization (-0. we include the share of Catholics in total population and an index of ethnolinguistic fractionalization.43). Political rights tend to be low in countries with high ethnolinguistic fractionalization ( population would be less prone to adopt jus soli elements. these stylized facts are in line with previous research and economic intuition.measured by the political rights variable . INSERT TABLE 5 Pairwise correlations among our dependent and independent variables are presented in Table 5.36). Current citizenship laws. The establishment of a consolidated democracy .35).should exert a positive effect on the probability of a jus soli legislation even though even in a democratic country hostility toward the assimilation of outsiders may persist for a protracted period of time. is highly correlated with the initial citizenship laws as identified by the jus sanguinis dummy (-0.64). the Catholic share. Pairwise correlations between all our independent variables are not reported for brevity and can be summarized as follows. In an effort to capture additional dimensions of cultural differences. Citizenship laws are also significantly 24 Page 26 correlated with migration. beside legal tradition. The Latin America dummy is positively associated with the Catholic share (0.31) and Latin America (-0. and ethnolinguistic fractionalization. while its correlation with the civil law dummy is much lower (-0. . thus displaying negative coefficients. Overall.

Results The results of our multinomial logits are presented in Table 6. which adds to (a) the dummies we discussed above. we find that the core variables are all significant. and that in the second subperiod the probability of applying a mixed or jus soli legislation increases. which adds to (b) the other potentially relevant economic and institutional regressors. migration and jus sanguinis display negative coefficients for both the mixed regime and jus soli. Jus sanguinis is the reference category for all the results shown. Hence. In specification (b).It is also clear that several of our independent variables are closely interrelated and that it may be difficult to disentangle their specific effect on the evolution of citizenship laws. Multinomial logit (a) is the core specification. namely. The period . relative to jus sanguinis. Multinomial logit (b) is an expanded specification. the jus sanguinis origin still exerts a negative impact on the probability of applying a mixed or jus soli regime. which includes the dummies. while migration 25 Page 27 only remains significantly negative for the probability of applying a mixed regime. migration and jus sanguinis as initial citizenship law. while both coefficients are positive for the period dummy. In particular. INSERT TABLE 6 Starting with the core specification (a). multinomial logit (c) is our full specification. 6. This means that high migration and a jus sanguinis origin decrease the probability of applying a mixed or a jus soli legislation rather than jus sanguinis. Finally. the results reported in the table indicate the impact of the explanatory variable on the probability of choosing either the mixed or the jus soli regimes. which includes only the core variables.2. The table reports three different specifications. plus the period dummy.

while the small country dummy now reveals a negative impact on the probability of applying a mixed regime. In this extended version the coefficient for decolonization loses significance. the decolonization dummy displays two negative coefficients. while the share of young in the population exerts a negative impact. in support of the hypothesis that countries with a relatively old population are more likely to choose mixed and jus soli regimes. The small country dummy is not significant in this specification. 15 The interaction term between migration and jus sanguinis origin is significant and positive for both the mixed regime and jus soli. in the full specification (c). while a high degree of democracy . Latin America has a positive coefficient for jus soli. the impact of migration. South Europe has a positive coefficient for the mixed regime. having gone through a decolonization border change negatively affects the probability of applying either a mixed or jus soli regime. However. confirming that Southern European countries have a higher probability of becoming mixed. namely. The size of government has a positive and significant coefficient for the probability of applying a jus soli regime.dummy is significantly positive only for the probability of a mixed regime. As expected. a jus sanguinis origin and the period dummy are confirmed. Finally. uncovering a tendency for countries with a jus sanguinis origin which are exposed to high migration to add jus soli elements. The share of Catholics and ethnolinguistic fractionalization are both insignificant. the strength of this tendency is questioned by the fact that. since these countries have a higher probability of applying this regime. The role of the Latin America and Southern Europe dummies is confirmed. the total effect of migration becomes negligible. if one evaluates the coefficients of migration and of the interaction together.

the marginal effect of 15 For both multinomial logit (a) and (b) we obtain the same results using a balanced sample composed by the 224 countries which constitute the reference sample for the estimation of our full specification. The marginal effects also allow to quantify the impact of our regressors.5%. having gone through a decolonization border change decreases the probability of being jus soli by about 14%. 17 The estimated marginal effects are calculated by holding all the independent variables at their mean 26 Page 28 decolonization reveals that this regressor retains a significantly negative impact on the probability of jus soli. 16 For all three specifications. Besides. while an increase in migration of one percentage point increases the probability of being a jus sanguinis country by about 2. In particular.3%. to be evaluated together with the 2. 17 Moreover.7%.positively affects the probability of applying either a mixed or jus soli regime. an increase in migration of one percentage point for a jus sanguinis country decreases the probability of being jus sanguinis by about 2. we verified that our results are not driven by outliers.5% increase due to direct effect of migration. inspection of the marginal coefficients in Table 7 confirms that migration increases the probability of jus sanguinis and decreases that of a mixed regime and that the interaction between jus sanguinis and migration is negative. . and decreases the probability of having a mixed regime by 2. 16 For the full specification (c). consistently with the positive coefficients for jus soli and a mixed regime in Table 6.

19 INSERT TABLE 7 Overall. Moreover. both per capita GDP and the Gini index of inequality tend to be associated with migration. 18 The correlation between Berlin wall and socialist is 0. 18 However.We also consider additional covariates that have often been found significant in related research on the determinants of institutions. and also with democracy and fractionalization. the original laws and our geopolitical dummies exert a significant impact on current citizenship laws.48. as well as the Berlin wall and the other border changes dummies. In fact they fail to add any further significance to the previous results. . Quantitative and qualitative development indicators such as income per capita and inequality could reveal if a richer. demographics and democracy also contribute to their determination. In levels. A dummy for socialist countries could instead work as an alternative to our Berlin wall border change dummy. For dichotomous independent variables the marginal effect is the change from 0 to 1 holding all other variables at their means. our results indicate that migration. more equal country is more prone to adopting jus soli elements. a dummy for oil countries could account for the fact that most of them have been experiencing huge immigration which has had no impact on their still very restrictive legislation (often based on Islamic family law). and that other factors such as government size. due to the fact that all countries identified by them do not exhibit enough variability with respect to the dependent variable. However. we cannot include in the regressions the socialist and oil dummies. so they are unlikely to add independent explanatory power to a regression.

The presence of countervailing forces highlighted by the model is therefore confirmed by our findings. the legal tradition interacts with the way countries react to migration in a complex way. we show that the legal tradition tends to affect the current legislation persistently. Overall. countries with a jus soli origin react to increasing migration by adding jus sanguinis elements. Therefore. our results suggest that migration pushes national legislations in the direction of jus sanguinis. for countries with a jus sanguinis origin which have experienced more immigration. in jus sanguinis countries the impact of migration turns out to be negligible. the welfare burden proves not to be an obstacle for a jus soli legislation. However. demo- . this correlation covers a more complex pattern which can be revealed once the interaction between migration and the legal tradition in the matter of citizenship is considered. However. On the other hand. Indeed. countries affected by Berlin wall have always applied jus sanguinis. not jus soli. because liberal countries tend to restrict while restrictive countries tend to resist innovation. this could be explained by the fact many of the countries with extended welfare systems may favor immigration because of their demographic crisis. On the one hand. as predicted by the model. the evidence does not the hypothesis of convergence toward a mix of provisions suggested by some political theories (see Section 1). there is evidence of a tendency toward adding jus soli provisions so that. even though the process of transplantation can prove discontinuous in the case of former colonies.19 For example. In addition. 27 Page 29 particular. Contrary to the model's implications. on balance.

3. Alternative Specifications We experiment our multinomial logit specifications with alternative covariates. relative open socialdemocracies. in particular as far as the impact of migration is concerned. we replace our measure of migration with a range of alternative measures. while cultural traits captured by religious affiliation and ethnolinguistic fractionalization appears to be irrelevant. 21 In the Table Appendix.graphic stagnation encourages the adoption of mixed and jus soli regimes. When we replace our migrations stocks with average migration flows (computed with reference to each subperiod). the impact we observe for the size of government could be explained by the fact that it proxies for European-style. our migration measure was chosen to minimize a potential endogeneity bias.1. and provide a deeper understanding of the forces shaping citizenship laws. This could suggest that migration flows are endogenous with respect to citizenship laws. In sum. 28 Page 30 6.3. 6. Moreover. 20 the coefficients for migration turns out to be insignificant in all three specifications. First. Robustness In this section we present a number of alternatives to our benchmark regressions to investigate whether they are robust to different specifications. samples and estimation techniques. . As outlined above. Further tests involve alternative measures of migration stocks. our empirical findings match the theoretical insights coming from the model. the evidence confirms that a higher degree of democracy is associated with more jus soli elements. Finally.

one could conclude that our detailed codification of the original citizenship laws does not add much to what we can already learn from a country’s broader legal tradition. column (2)). Results from this instrumented specification are similar to those presented in Table 6.45. 22 Since jus sanguinis and jus soli are in principle closely linked to the civil and common systems of laws. as possible alternatives. Since in the postwar period migration has been highly regulated by policy in most receiving countries.Table A. the simultaneous determination of citizenship laws and migration does represent a concern when we enter within-the-period data on migration instead of beginning-of-period data. for the first and the second subperiod we select the years 1970 and 1990. respectively. 23 If the coefficients of the two alternative dummies were the same.1.1. Some differences emerge in the coefficients for migration and its interaction with jus sanguinis. and citizenship laws could be viewed as part of migration policy. and the 1960-70 and 1980-90 averages. the latter turns out to be substantially less significant and reduces the 20 Migration stocks and flows show a correlation of 0. with the latter losing significance. respectively. 21 Namely. our beginning-of-period migration stocks prove to be the most adequate measures of the role of migration. the influence of the legal tradition can also be analyzed through a dummy capturing the presence of a civil law tradition. . column (1) presents the full specification with the 1960-70 and 1980-90 average migration stock. 24 When we replace the jus sanguinis with the civil law dummy (see Table A. To sum up. 22 We also experiment with a specification entering the migrant stock in 1960 for both subperiods. respectively.

we assume that jus sanguinis corresponds to minimal and jus soli to maximal inclusiveness. In particular. Our previous conclusions . Alternative Estimation Techniques Alternative estimation techniques broadly confirm the same results from Table 6.2 and A. When we replace the Southern Europe dummy with migration flows. The correlation between civil law and jus sanguinis is 0. 6. we find that its coefficient is insignificant.2. These results are reported in Table A. 29 Page 31 significance of most other regressors. an (unreported) alternative multinomial probit model delivers the same qualitative results.35.3. which we assume here to be ordered by increasing inclusiveness.35. We also replace our decolonization dummy with a dummy for British or Portuguese colonies (identifying those countries that were characterized by a jus soli legislation during the colonial period). suggesting that civil law is a much weaker predictor of current citizenship laws than the original citizenship laws. namley. and with the subSaharan Africa dummy. since it is the quickly increasing second-subperiod immigration which determines the peculiar behavior of this region. 24 The correlation between the two dummies is 0. We also run ordered logit regressions where current citizenship laws are explicitly treated as an ordinal variable. possibly because flows are multicollinear with respect to stocks and endogenous with respect to the dependent variable. 25 but both alternatives are associated with insignificant coefficients.3.23 The dummy is equal to 1 if a country belongs to the civil law tradition and to 0 if a country belongs to common law.

Alternative Sample Criteria We also run multinomial logit regressions on a cross-sectional sample composed by country averages over the 1950-2000 period. namely. 27 The results for this cross section reveal a much lower level of significance for several covariates. the test provides evidence that the parallel regression assumption has been violated. 26 For all specifications.31). 30 Page 32 with the initial laws. an alternative ordered logit regression also achieves much weaker results than in the panel. In all cases.30) and for sub-Saharan Africa (0.are confirmed.3. with migration and a jus sanguinis origin exerting a negative impact on the application of jus soli. The test is an approximate likelihood-ratio test of proportionality of odds across response categories. Our multinomial logit specification is hence superior to an ordered logit specification. while migration stock refers to 1960 and jus sanguinis in 1948 is the initial law. 1950-1975 and 1976-2000. 28 The same applies to alternative variants with a cross section over each subperiod. For the same cross section. 27 Here the dependent variable is citizenship laws in 2001. in particular for migration and for its interaction 25 Decolonization shows a significant correlation with the dummies for British or Portuguese colony (0. 26 6. 29 while ordinary least squares regressions with our citizenship policy index in 2001 and an indexed version of . we run a test for the parallel regression assumption.3.

4).1. 30 These results may be due to the fact that a single cross section of countries includes less information than our panel. (10) where α is a constant term and η it is an error term. 0 if no change occurs.2001 citizenship laws as alternative dependent variables both show an insignificant coefficient for migration (see Table A. the dependent variable is categorical and can take three values: -1 if the country changes its laws toward jus sanguinis. Empirical Specification In the alternative specification. we run a multinomial logit model of the following form for its full specification: V it = α + βM it + γT t +Z 0 it δ+η it . 7. 7. While specification (9) focuses on current citizenship laws as the dependent variable. For our panel. Z it contains all the variables previ- . we developed an alternative specification which is designed to capture more specifically the determinants of a change in the laws. An Alternative Approach: The Determinants of Change in Citizenship Laws In this section we study citizenship laws evolution using an alternative approach which is able to provide additional insights. and 1 if the country changes its laws toward jus soli.

ously described. jus sanguinis as the initial citizenship law. 31 Page 33 the independent variables are presented in Table 5. namely. showing that migration is again negative correlated with the dependent variable. and with average migration flows. with similar results. Pairwise correlations among our new dependent variable. 29 We apply to the above results the Cook's distance method. The size of government's negative coefficient . except for those which present zero-cell problems. the results reported in the table indicate the impact of the explanatory variable on the probability of choosing either restriction toward jus sanguinis (first column) or expansion toward jus soli (second column).2. its interaction with migration. Migration has a positive impact on the probability to restrict and a non significant impact on the probability to expand. to show that they are not driven by outliers. Decolonization and Southern Europe exert a positive impact on change in both directions. where the reference category is no change. 7. Hence. changes in citizenship laws. while the opposite holds for Southern Europe. more significantly so for restriction for the case of decolonization. and 28 We also experiment with migration stocks in 1970 and 1980. The period dummy indicates that the second subperiod witnesses an increase in the probability to expand. 30 The same qualitative results arise in an unreported regression with naturalization in 2001 as dependent variable. Results Regression results for the multinomial logit model are presented in Table 8. and the Latin America dummy.

Ethnic diversity emerges as a significant factor of change toward jus sanguinis. 32 Page 34 impact on expansion. Alternative measures of migration confirm an irrelevant impact on change in the laws. Overall. due to the removal of the Latin America dummy which comprises several former Portuguese colonies. we perform a full set of robustness checks for (10). in Table 9. The Gini index once again fails to add any significance.for restriction confirms that this factor actually prevents it. the implications of Table 9 are in line with those of Table 7. while again a relatively young population provokes resistance to extension. 31 Moreover. The subSaharan Africa dummy is not significant when entered instead of decolonization. in Table 7 decolonization exerts a significantly negative impact on the probability of having a jus soli legislation. INSERT TABLE 8 Table 9 reports the marginal coefficients for the regressions in Table 8 and confirms the restrictive impact of migration emerging from Table 8. which reports the marginal effects for the multinomial logit specification. these results are complementary to those of Section 6 since they highlight which . while per capita GDP reveals a significant 31 The estimated marginal effects are calculated as explained in footnote 17. which is consistent with the fact that. the alternative dummy is again significant. meaning that relatively old countries are more likely to liberalize their legislation. it emerges as a factor that facilitates change toward sanguinis. When we replace our decolonization dummy with the dummy for British or Portuguese colonies. INSERT TABLE 9 As for specification (9). For instance.

We developed a simple median voter model where citizenship rights are granted by natives to migrants on the basis of the associated benefits and costs. on the basis of a new dataset we compiled. the legal tradition has affected the way countries have responded to migration. we found that indeed citizenship laws have responded endogenously and systematically to a number of economic and institutional factors. Countries with larger welfare systems. Moreover. while in jus sanguinis countries the impact of migration has been negligible. Border instability emerges as a decisive factor in shaping citizenship laws.factors have induced the observed evolution of the legislation. reflecting discontinuities for the transplanting process of legal institutions. migration is confirmed as a factor that favors change toward restriction. In particular. the evidence does not support the hypothesis of convergence toward a mixed regime that includes both jus soli and jus sanguinis elements. Conclusion We studied the theoretical and empirical determinants of the legal institution of citizenship in the postwar period. In particular. jus soli countries have reacted to increasing migration through restriction. Therefore. In bringing the model to the data. Our investigation reveals that migration has had an overall negative impact on liberalization of the legislation and the adoption of jus soli elements. older population and more extended political rights tend to be associated with more diffused elements of jus soli. particularly in connection with the decolonization phase. 8. . for each possible direction of change. The model predicts that migration has a potentially ambiguous impact on the legislation and that this impact is also affected by cultural factors including a country's degree of inclusiveness. as reflected by the original citizenship legislation.

have already been investigated. citizenship laws can be viewed as a link. Many issues that fall within the former . The Citizenship-at-birth Classification Group 1 (jus sanguinis countries): We include countries where citizenship is passed on to a child based upon at least one of the parents being a citizen of that country. Our methodology can be extended to the study of other evolving bodies of the law. The endogeneity of institutions to economic factors represents a challenge for research aimed at demonstrating that institutions are crucial determinants of economic performances. In the application of jus sanguinis. by using projections of international migration patterns in combination with the available predictions about the future course of democratization and border changes. a clear implication of our investigation is that institutions should not be 33 Page 35 presumed to be exogenous.such as commercial law. By showing that citizenship laws are shaped not only by the broader legal origins. Finally. but also by other institutions such as the internal system of political rights and the international system of relations as reflected by state borders. Further research will study the future evolution of citizenship policy. regardless of the child's actual country of birth. such as family law. between the public and the private sphere of influence. within a legal system. rules of inheritance. Citizenship laws are still changing. we also establish that different institutions are interrelated. since they do adapt both to economic and non-economic factors. Data Appendix A. countries may differ on some .More generally. and where citizenship is not granted due to birth within the country. labor regulation. and women's rights. and government activities . This represents another challenge for further research on the process of formation of legal rules and on the impact of institutions on economic outcomes.

For example. Group 2 (countries with a mixed regime): We include those countries where elements of jus soli are recognized. automatic citizenship for the children of those immigrants who were also born in the country). for example on the father's vs. regardless of the parents' citizenship or status. Another is the ability. . the existence of a provision that birth in the country matters for naturalization. the requirement of citizenship for one or both parents. Moreover. and the requirement that parents must be citizens other than by descent. that justifies the inclusion of a country within Group 2. Group 3 (jus soli countries): We include those countries where citizenship is automatically granted due to birth within the country. and coexist with varying degrees of jus sanguinis. the relevance of the marital status of the parents. residence requirements for parents. Examples of restrictions are generational requirements limiting the principle of citizenship by descent to the first or second generations of individuals born and residing abroad. Since we focus on the presence of jus soli elements in a country's legislation. we interpret as an element of jus soli. Most of these factors depend on the interaction between local family law and citizenship law. A common exception to the general principle of jus sanguinis is automatic citizenship attribution to children of unknown parents.factors. a frequent provision that limits jus soli is double jus soli (namely. albeit in a restrictive form. mother's right to transmit citizenship by descent. the age of maturity) subject to either residence requirements or application. for a child born in a country were jus sanguinis prevails. to acquire citizenship at some later point (for example. our classification does not emphasize how narrowly 34 Page 36 jus sanguinis can be specifically applied to emigrants.

the State Transformation of Germany in the postwar decolonization phase. We refer to Bertocchi and Strozzi (2009) for further detail on the data set on citizenship laws. and the USSR in 1991. residence requirements). Clearly. there is substantial overlap among the observations recorded in the Polity IV dataset. and Russia in 1992). from the Polity IV variable CHANGE we record information on four types of events capable of affecting state borders. We . In particular. West Germany in 1945. State Demise. Even if we set 1948 as the initial date for our citizenship laws analysis. State Disintegration. West Germany in 1990. Decolonization. for example. plus a few additional observations not linked to these two waves. namely. East Germany in 1945. State Transformation. and State Creation. the new countries formed in Europe after the fall of the Berlin wall. for border changes we include a few earlier events occurred in the 1943-1948 period that fit within the phase of post-colonial independence. The Border Change Dummies We construct three border change dummies (namely. Berlin Wall. Examples of the events contained in the Polity IV 35 Page 37 (2002) dataset are the State Disintegration of Yugoslavia in 1991. The countries affected by State Creation are the most numerous. the State Demise of Germany in 1945. B. They include the new countries gaining independence . East Germany in 1990. and Other Border Changes) based on data collected from Polity IV (2002).Normally countries that apply jus soli combine it with jus sanguinis provisions for the children of their citizens born outside of their territory (although limitations to the ability to transmit citizenship acquired in this manner to the next generation usually apply through.and therefore state borders .

For instance. Likewise. was obtained from the CIA (2002). The data are taken from United Nations (2003) and are available for 1960. The data refer 36 Page 38 to incoming international migrants less outgoing international migrants. the separation of Bangladesh from Pakistan counts for two events. 1970. we count as a single event.000 total . Net migration flows: International net migration rate. but also the State Demise of Germany in the same year. and the unification of Vietnam). On this basis. the absence of border changes. occurring to Germany. 1980. including refugees. On the other hand. namely. Additional information. we treat as another single event. Migration stock is the number of people born in a country other than that in which they live. To be noticed is that the way our variables are coded reflects stability of borders. more than the direction of a change in terms of a country's size. when necessary. occurring again to Germany. Berlin wall (identifying countries which went through a post-1989 Berlin wall border change). per 1. 1990 and 2000. we construct our three border change dummies for each period under consideration: Decolonization (identifying countries which went through a post-colonial redefinition of their borders). the State Transformation of Germany in 1990 and the State Demise of East and West Germany in the same year. of which examples are the split between Pakistan and Bangladesh. the State Transformation of East and West Germany in 1945. Definitions and Sources of Other Covariates Migration stock: International migration stock (% population).adapt these data to our needs by matching them to the 162 countries appearing in our citizenship laws dataset. C. because it concerns two countries which are in our sample. and other border changes (identifying countries which went through other types of boundary changes.

Latin America. Southern Europe and sub-Saharan Africa: Dummies for countries belonging to Latin America. and Brunei). without distinguishing. Southern Europe and subSaharan Africa. Small country: Dummy for countries with a population size of less than one million over all available years between 1960 and 1995. Catholic share: Percentage of Catholics in 1980. GDP per capita: Logarithm of real GDP per capita at current international prices. Government consumption: Government share of GDP in current prices. Share of young: Share of young between age 15 and 34 (% population). Information is from La Porta et al. with projections until 2050. Civil law: The source is the legal origin classification in La Porta et al. The source is the Correlates of War 2 Project (2004). Oil: Dummy for oil countries (OPEC countries plus Oman. taken from Freedom House (1996). (1999). Bahrain. Ethnolinguistic fractionalization: Composite index of ethnolinguistic fractionalization. since they do not present any significant difference for the issue of citizenship. taken from La Porta et al. taken from Easterly and Levine (1997). The data are available over five year intervals from 1950. taken from Penn World Tables (2002). while La Porta et al. Political rights: Political rights index. Qatar. Socialist: Dummy for socialist countries. (1999). We retain only the two main families of common and civil law. INSERT TABLE APPENDIX References 37 . The classification is from UN (2002). among the French. The source is the United Nations (2005). Gini index: Gini index of inequality. (1999). taken from Deininger and Squire (1996). Angola. Moreover. within the broader civil law tradition. as in Easterly and Kraay (2000). and Scandinavian versions. taken from Penn World Tables (2002). The data are taken from United Nations (2005). we assign them to their own class of common or civil law as it prevailed before the communist period. German. (1999) introduce a separate class for socialist-law countries. British or Portuguese colony: Dummy for countries that were British or Portuguese colonies any time after 1918.population.

Washington: Carnegie Endowment for International Peace. Aleinikoff. Acemoglu. Graziella. Klusmeyer (edited by). National Bureau of Economic Research. 2000. European Economic Review 40:1737-1744. Rainer. Bertocchi. and Douglas B. Florencio Lopez-de-Silanes.Page 39 Acemoglu. 2001. 2008. Kees Groenendijk. Bertocchi. The Divergence of Legal Procedures. Cambridge. Working Paper no. Balas. Jess. The Politics of Co-optation. 2007. On the Number and Size of Nations. Centre for Economic Policy Research. Journal of Comparative Economics 29:591-607. London. Citizenship Today: Global Perspectives and Practices. Inequality and Growth in Historical Perspective. Alesina. On the Political Economy of Migration. Klusmeyer (edited by). Rafael La Porta. and Andrei Shleifer. Alberto. Quarterly Journal of Economics 115:1167-99... Mass. Amsterdam: Amsterdam University Press. American Economic Review 91:1369-1401. The Colonial Origins of Comparative Development: An Empirical Investigation. 1997. Thomas A. Daron. 6396. From Migrants to Citizens: Membership in a Changing World. and Michael Spagat. Daron. Discussion Paper no. Aron. Eva Ersbøll. Washington: Carnegie Endowment for International Peace. Thomas A. Robinson. Graziella. 2006. Aleinikoff. Benhabib. 2000. and James A. 38 Page 40 . Simon Johnson. The Enfranchisement of Women and the Welfare State. Acquisition and Loss of Nationality. Why Did the West Extend the Franchise? Democracy. and James A. Quarterly Journal of Economics 112:1027-1056. Baubock. 13809. and Harald Waldrauch (edited by). 1996. and Enrico Spolaore. and Douglas B. 2001. Robinson. 2001.

3596. Journal of Immigration and Integration 6:435-468. and Chiara Strozzi. 2008.. The Economic Causes and Consequences of Canadian Citizenship. 2004. Immigration Policy: Methods of Economic Assessment. 2000. Growth. The World Bank Economic Review 10:565-91. Quarterly Journal of Economics 112:1057-1090. Cambridge: Harvard University Press. 2002.unimore.. Don J. William. Department of Political Science. Public Choice 137:81-102. Small Problems? Income. Bolton. 1997. International Economic Review 45:1129-1168. Bertocchi. British Council Klaus. Citizenship and Nationhood in France and Germany. Miller. and Lyn Squire. DeVoretz. The Breakup of Nations: A Political Economy Analysis. Barry R. DeVoretz. Bonn. 1992. 1996. International Migration and the Role of Institutions. and Volatility in Small States. Brubaker. Graziella. World Development 28:2013-2027. Africa's Growth Tragedy: Policies and Ethnic . and Sergiy Pivnenko.Bertocchi. 2006. Citizenship in the United States: The Roles of Immigrant Characteristics and Country of Origin. European Civic Citizenship and Inclusion Index. and Ross Levine. Small States. Deininger. Don J. and Chiara Strozzi. IZA. 1997. 2009. Huffman. Brussels: British Council Brussels. Correlates of War 2 Project. 39 Page 41 Easterly. Central Intelligence Agency. Easterly. Patrick. On the Political Economy of Migration and Income Redistribution. Dolmas. The World Factbook 2002. William. The Citizenship Laws Dataset. and Gerard Roland. Washington: CIA Chiswick. Discussion Paper no. The Pennsylvania State University. and Paul W. Graziella. Jim.asp.recent. 2004. Rogers. A New Data Set Measuring Income Inequality. International Migration Review 40:390-418. 2008. 2006. and Gregory W. and Aart Kraay. 2005. http://www.

Langhammer. edited by Federico Foders and Rolf J. Samuel P. 2006. and Bettina Aten. Timothy J. Andrei Shleifer. 2004.. Rafael. Mass. Quarterly Journal of Economics 112:1203-1250. Mark. and Paths of Development Among New World Economies. Heidelberg: Springer. 2006. Anna Maria. 1999. Journal of Population Economics 19:32744. Engerman. Freedom in the World. 2006. 2002. Journal of Law. La Porta. and Maurice Schiff.Divisions. Vishny. Cambridge. Heston. 1998. Working Paper no. Florencio Lopez-de-Silanes.. Penn World Table Version 6. Alan. Sokoloff. Who is Against Immigration? A Cross-country Investigation of Individual Attitudes towards Immigration. National Bureau of Economic Research. and Kenneth L. Inequality. La Porta. The Quality of Government. and Jeffrey G. Political Rights and Civil Liberties. Review of Economics and Statistics 88:510- . New York: Freedom House. Gradstein. Economics and Organization 15:222279. October. Law and Finance. New York: Simon & Schuster. 1-34 in Labor Mobility and the World Economy. Rafael. Negative Selection and Policy. Oxford: Oxford University Press. Vishny. Freedom House. and Robert W. The Political Economy of Social Exclusion with Implications for Immigration Policy. 1998. Center for International Comparisons at the University of Pennsylvania (CICUP). Challenge to the Nation-State: Immigration in Western Europe and the United States.1. Williamson. and Robert W. Journal of Political Economy 106:1113-1155. International Migration in the Long Run: Positive Selection. Stanley L. Joppke. Christian (edited by). Hatton. Factor Endowment. 2002. Robert Summers. 40 Page 42 Mayda. Pp. Florencio Lopez-de-Silanes. Huntington. 1996. Who Are We: The Challenges to America's National Identity. Andrei Shleifer. 9259.

Polity IV. Investigations Service.530. Efraim Sadka. United Nations. and Jeffrey G. 2006. 1800-2002. and Globalization Backlash. Alan H. Trends in Total Migrant Stock: The 2003 Revision. Geneva: United Nations. High Commissioner for Refugees.. United Nations. 2003. Thomas A. Kevin H. European Journal of Political Economy 22:838-861. Access to Citizenship: A Comparison of Twenty-Five Nationality Laws. 2002. 2005. O'Rourke. Policy Interaction. and Phillip Swagel. United States Office of Personnel Management. 41 Page 43 Weil. Lant. Pritchett. Klusmeyer. Ashley S. Common Database. Williamson. Washington: The Brookings Institution Press. Immigration Policy Prior to the Thirties: Labor Markets. Population and Development Review 24:739-771. NewYork: United Nations. Meltzer. Journal of Public Economics 85:167-190. New York: United Nations. United Nations. Richard.. Tax Burden and Migration: A Political Economy Theory and Evidence. 2006. Let Their People Come: Breaking the Deadlock in International Labor Mobility. Patrick. 2002. The Determinants of Individual Attitudes towards Immigration. Razin. Citizenship Laws of the World. World Population Prospects: Population Estimates and Projections. and Scott F. Assaf. 42 Page 44 Table 1 C . 1998. edited by Aleinikoff. Washington: Carnegie Endowment for International Peace. Political Regime Characteristics and Transitions. University of Maryland. 1981. 2001.17-35 in Citizenship Today: Global Perspectives and Practices. and Douglas B. 2003. Timmer.. A Rational Theory of the Size of Government. and Richard Sinnott. Pp. Country of Origin and Legal Information. 2002. Journal of Political Economy 89:914-27. New York: United Nations. 2001. Center for International Development and Conflict Management. United Nations.

ITIZENSHIP L AWS E VOLUTION :T RANSITION M ATRICES Citizenship laws in 2001 Jus sanguinis regime Mixed regime Jus soli regime Total Citizenship laws in 1948: Jus sanguinis regime 46 20 1 67 Mixed regime 11 6 2 19 Jus soli regime 31 9 36 76 Total 88 35 39 162 Citizenship laws in 1975 Jus sanguinis regime Mixed regime Jus soli regime Total Citizenship laws in 1948: .

Jus sanguinis regime 63 3 1 67 Mixed regime 10 7 2 19 Jus soli regime 28 1 47 76 Total 101 11 50 162 Citizenship laws in 2001 Jus sanguinis regime Mixed regime Jus soli regime Total Citizenship laws in 1975: Jus sanguinis regime 81 20 0 101 Mixed regime 2 9 0 11 Jus soli regime 5 6 39 50 .

Total 88 35 39 162 Page 45 Table 2 C HANGES IN C ITIZENSHIP L AWS Changes in citizenship laws (1948 to 2001) No change Toward jus sanguinis Toward jus soli Total Citizenship laws in 1948: Jus sanguinis regime 46 0 21 67 Mixed regime 6 11 2 19 Jus soli regime 36 40 0 76 Total 88 51 23 162 Changes in citizenship laws (1948 to 1975) No change .

Toward jus sanguinis Toward jus soli Total Citizenship laws in 1948: Jus sanguinis regime 63 0 4 67 Mixed regime 7 10 2 19 Jus soli regime 47 29 0 76 Total 117 39 6 162 Changes in citizenship laws (1975 to 2001) No change Toward jus sanguinis Toward jus soli Total Citizenship laws in 1975: Jus sanguinis regime 81 0 20 101 Mixed regime 9 2 0 11 Jus soli regime 39 11 .

0 50 Total 129 13 20 162 Page 46 Table 3 C ITIZENSHIP L AWS D ATA S ET :S UMMARY S TATISTICS Variable Observations Mean Standard deviation Minimum Maximum Citizenship laws in 2001 162 1.698 .685 .835 1 3 Citizenship laws in 1975 162 1.916 1 3 .

920 1 4 Citizenship policy index in 2001 142 .056 .173 .204 .655 -1 1 Naturalization in 2001 142 2.941 1 3 Changes in cit. laws (1948 to 1975) 162 .Citizenship laws in 1948 162 2. laws (1975 to 2001) 162 -.451 -1 1 Changes in cit.488 -1 1 Changes in cit. laws (1948 to 2001) 162 -.458 .415 .299 0 1 Page 47 Table 4 S UMMARY .043 .

080 .673 Net migration flows 318 -.691 .500 0 1 Civil legal origin 324 .485 -1 1 Jus sanguinis as initial cit.713 .646 9.519 .011 70.S TATISTICS Variable Observations Mean Standard deviation Minimum Maximum Current citizenship laws 324 1.470 0 1 Migration stock 300 5. law 324 .875 1 3 Changes in citizenship laws 324 -.673 .116 .

022 .250 .052 .379 0 1 Southern Europe 324 .160 .173 .368 0 1 Sub-Saharan Africa 324 .551 -47.434 0 1 Berlin wall 324 .35 Decolonization 324 .95 63.8.043 .146 0 1 Latin America 324 .223 0 1 Other border changes 324 .204 0 1 Small country 324 .

173 .281 0 1 Political rights 276 3.333 .572 0 97.379 0 1 Oil 324 .299 0 .535 35.265 .3 Ethnolinguistic fractionalization 272 .890 Government consumption .730 2.086 .349 ..040 1 7 Catholic share 324 31.442 0 1 British or Portuguese colony 324 .472 0 1 Socialist 324 .

249 4.263 19.966 20.495 63.425 9.01 53.54 3.979 10.180 N OTE .617 11. Page 49 Table 5 P AIRWISE C ORRELATIONS A .233 Share of young 324 34. For details about the construction of the variables see the text.47 27.492 72. The first cross section refers to the 1950-1975 subperiod while the second cross section refers to the 1976-2000 subperiod.488 1.342 2. The sample includes two cross sections of 162 countries. The reference period is 19502000.36 Page 48 Log GDP per capita 263 7.060 Gini index 155 40.

12* .08 .60** .33** Decolonization -.02 Latin America .MONG D EPENDENT AND I NDEPENDENT V ARIABLES Citizenship laws Changes in citizenship laws Migration stock -.64** .05 .42** Berlin wall -.48** Civil legal origin -. law -.07 -.14* -.09 -.11 + -.24** Other border changes -.10 + Small country .12* -.15** .06 Jus sanguinis as initial cit.12* Net migration flows -.04 Southern Europe -.

34** Gini index .03 .06 Political rights .05 .12* -.Sub-Saharan Africa -.02 -.18** Catholic share .22** -.24** Page 50 N OTE . .14* Oil -.26** Socialist -.13* -.10 + Ethnolinguistic fractionalization -.25** . The reference period is 1950-2000.03 -.29** .06 Share of young -. The sample includes two cross sections of 162 countries.39** .10 + -.20** Log GDP per capita .19** British or Portuguese colony .20** Government consumption -.

051 + -.10. + p < .055** . For details about the construction of the variables see the text.01.05.The first cross section refers to the 1950-1975 subperiod while the second cross section refers to the 1976-2000 subperiod. ** p < . Page 51 Table 6 T HE D ETERMINANTS OF C ITIZENSHIP L AWS :M ULTINOMIAL L OGIT E STIMATES Specification (a) Specification (b) Specification (c) Mixed regime Jus soli regime Mixed regime Jus soli regime Mixed regime Jus soli regime Migration stock -. * p < .

61) (-2. cit.86) (-2.53) Jus sang.24) (-4.296 + -2.27) Decolonization -1.048** -2.30) (-1.032 -.90) (-5.39) (-5.55) . law -2.65) (.887** 1.211 (4.523 (-1.59** (-4.272 2.09) (-1.385** .712** -6.117** -4.28) (3.43) (-4.084 -1.02) (2.47) (2.193** 1.41) (-.568** -7.054 + -.79) (-1.167* -.211** -.38) Period 1.81) (-1.044** -6.28) (-.87) (-3.045 (-1.-. as init.409* -.99) (-4.

234 + 1.495** (-.81) (1.14) (1.156 + .523 (-1.88) (.03) (-.415 -2.23) (.866** -.153 1.70) (3.804 .063** (1. X migration stock .224* -.035 .187 (2.07) (2.Southern Europe 1.68) .621 4.566 + .75) Government consumption .69) (.93) Small country -.64) Jus san.799 2.51) (2.73) (-.87) Latin America -.69) (-1.084 + (1.

426* (2.626* -.09) N 300 300 300 300 .397* (-1.70) Ethno.84) (2.165* (-2.548** .43) Page 52 Catholic share .716 11.01 (1.02) Political rights .023 1.193 + -.Share of young -.02) (1.092** .353* 3. fractionalization .024 1.012 .277 (.03) (2.73) (2.46) (.05) (2.75) (-2.48) (-.43) (1.46) Constant -1.

24 Maximum Likelihood R 2 .77 -155.224 224 Log likelihood -195.32 .94 -95.79 Count R 2 .45 .68 McFadden's R 2 .29 .55 . clustered at country level.57 McFadden's Adjusted R 2 .4 . The reference period is 1950-2000.44 Cragg & Uhler's R 2 . . Robust z statistics in brackets.81 .64 N OTE .58 .32 .53 .83 Adjusted Count R 2 .46 .68 . Jus sanguinis is the reference category.71 .

681** (18.001 (2.7) . * P < .875** -.34) Jus sanguinis as initial cit. law .09) (-.193* -.F ULL S PECIFICATION ) Jus sanguinis regime Mixed regime Jus soli regime Migration stock .023* -.10. Page 53 Table 7 T HE D ETERMINANTS OF C ITIZENSHIP L AWS :M ARGINAL E FFECTS (M ULTINOMIAL L OGIT .01.05.025* -.+ P < .49) (-2. ** P < .

51) Latin America -.088 -.016 -.123 .034 (-1.170 .114 (.121 (-1.77) Southern Europe -.86) (-.194** .85) (5.03) (.59) (-3.80) (.51) (1.16) (.44) Decolonization .291 .027* .92) Period -.56) (-7.805** (-4.139 + (.203** .303 + .95) (2.(-2.75) Small country .86) Jus sanguinis X migration stock -.336** -.13) (-1.611** -.15) (-2.

.01) Share of young .028 .021 .002 .92) Government consumption -.15) Catholic share -.006 (-2.50) .069** .36) (1.004 .152 (-.50) (.29) (1.001 Page 54 (-1.018 -.124 -.007* (-1.56) (-.001 . fractionalization -.16) (.061** -.96) Political rights -.76) (2.18) (-2.107** .043* (2.98) (.65) (-1.037 + (-3.010* .50) Ethno.14) (.

The above marginal effects refer to the multinomial logit estimates of our full specification (specification (c) in Table 6).90) N 224 224 224 N OTE . ** P < . Robust z statistics in brackets.61) (1.(2. clustered at country level.039 + .05. + P < .01. The reference period is 1950-2000. Page 55 Table 8 T HE D ETERMINANTS OF C HANGE IN C ITIZENSHIP L AWS :M ULTINOMINAL L OGIT E STIMATES Toward jus sanguinis Toward jus soli Migration stock . * P < .10.

65) (1.001 (.155** (1.-.006 .94) (3.12) Share of young .73) Decolonization 2.44) Small country -.05 + .898** 2.058 + (2.593 (.15) (2.82) (1.493** (.68) Southern Europe 2.2 + 2.90) (.12) Period .50) (.652 1.82) Catholic share -.70) .60) Government consumption -.95) (1.032 (1.02 -.736 + (.006 (1.341 -.

508 (2.96) (2.23) (1.58 Maximum Likelihood R 2 .04) N 224 224 Log likelihood -107.364 9.08) Constant -4.84) (1.33 McFadden's R 2 .12) Ethnolinguistic fractionalization 1.16 .3 McFadden's Adjusted R 2 .45 Count R 2 .52) Political rights .771* Page 56 (.104 .81 Adjusted Count R 2 .(.234 (.15 Cragg & Uhler's R 2 .981* 2.

26) . Page 57 Table 9 T HE D ETERMINANTS OF C HANGE IN C ITIZENSHIP L AWS :M ARGINAL E FFECTS (M ULTINOMIAL L OGIT ) Toward jus sanguinis No change Toward jus soli Migration stock . clustered at country level.N OTE . + P < . No change in citizenship laws is the reference category.000 (2.01.01) (-1. The reference period is 1950-2000.05.10. * P < . ** P < . Robust z statistics in brackets.003* -.003 -.

06) Small country -.016** (.28) Share of young .015 (-.00) (-2.069 (2.19) Period .013 -.16) Southern Europe .058* (.53) (.58) (-1.(-.33) (1.116 (1.004* .022 .442* .299 -.003 .65) Government consumption -.05) Decolonization .74) (-.10) (-2.003 .31) (2.415* .044 -.42) (1.001 (-2.038 -.27) (1.102 .29) .02) (1.374* -.

007 -. + P < .10.75) (-1.1 .10) (1.74) Catholic share -. Page 59 Table Appendix Table A.(1.12) N 224 224 224 Page 58 N OTE .223* .89) Political rights . clustered at country level.000 .05.015 .18) (-2.60) (.12) (-2.71) (. * P < .000 (-.077 + (2. ** P < .145* -.19) Ethnolinguistic fractionalization .000 .007 (.01.39) (1. The above marginal effects refer to the multinomial logit estimates in Table 8. The reference period is 1950-2000. Robust z statistics in brackets.

62) Jus sanguinis as initial citizenship law -4.36) Average migration stock -.T HE D ETERMINANTS OF C ITIZENSHIP L AWS :M ULTINOMIAL L OGIT E STIMATES .A LTERNATIVE C OVARIATES Specification (a) Specification (b) Mixed regime Jus soli regime Mixed regime Jus soli regime Migration stock -.344** .152* -.056 -.052 (-2.533** -7.013 (-.81) (-.12) (-1.

31) Decolonization -.76) (1.218 1.908 .525 1.003 (-.941 -1.42) Latin America -.834** -.09) (-4.317** -.397** .84) (-1.742 + (1.64) (.649 (2.260 1.79) (-.(-5.88) Period 2.21) (-1.22) (.16) (-1.68) (.667 4.12) Civil legal origin .046 5.595 (1.062 .140 -1.01) Southern Europe 1.509 + -.33) (2.436** .54) (1.

402* .64) Small country -2.67) (-1.128 (-1.040 Page 60 (1.110 (.73) (-.61) Civil legal origin X migration stock .197 + -.036 .140 + .03) (3. X average migration stock .503 + .068** (1.23) (-1.560 -1.89) (1.65) (1.90) (.29) Jus san.25) (2.894 + 1.55) (2.65) (.10) Government consumption .09) (2.(-.063** .019 -.59) Share of young -.043 .

03) (.011 .096 -.527 (1.831 11.11) .017 (1.52) (.436* .029 1.08) (2.46) (-1.56) Constant 3.035 1.75) (.290 1.98) Catholic share .012 .006 .247* (-1.563* .16) (-1.950 + (-.567 4.78) (-2.03) (1.81) (1.38) Ethnolinguistic fractionalization -.58) (1.74) Political rights .276 (2.539** .85) (1.78) (2.-.52) (2.273* -3.

Robust z statistics in brackets. ** P < .57 .10.(-1.41 McFadden's Adjusted R 2 .83 . .28 Cragg & Uhler's R 2 .78 .44 .01 Maximum Likelihood R 2 .05.01) N 224 224 224 224 Log likelihood -95. Jus sanguinis is the reference category.04) (1. The reference period is 1950-2000.55 McFadden's R 2 .75 Adjusted Count R 2 .01.74 -132. * P < .48 N OTE .68 . clustered at country level.64 Count R 2 . + P < .64 .

Page 61 .